Hacker News new | comments | show | ask | jobs | submit login
LaTeX Coffee Stains (2009) (hanno-rein.de)
279 points by gandalfar 343 days ago | hide | past | web | favorite | 36 comments



Some earlier research on this topic:

http://arxiv.org/abs/0903.3365

and

http://arxiv.org/abs/0812.3367


any idea as the to the purpose?

are the stains preprint(o) identifiers?

or just good humored researchers?

or troll(i) identifiers like bad grammar often is?

or some undiscussed other?

(o) http://www.astrobetter.com/blog/2011/12/12/to-post-or-not-to...

(i) http://snarxiv.org/vs-arxiv/


wrong thread?


Have a quick scan through the papers...


Thanks. Got it.


Along similar lines, I'm a fan of the "latex-therefore" package: https://github.com/bgschiller/latex-therefore


I looked for something similar but with red wine instead of coffee for my PhD thesis. It didn't exist so in the end I just used a scanned stain of red wine and the wallpaper package :).


There is now a wine stains version too :) https://www.overleaf.com/latex/examples/latex-wine-stains/cd...

(although I like your workaround!)


This is right up there with \usepackage{amsmath} as a truly must-have package for any real scientific paper.


Excellent. I've been looking to write some papers in heptapod, and found that LaTeX was woefully not up to the task. This looks like a great starting point for future expansion; thanks!


This could create a whole digital market for food stains. Mac&Cheese, soy sauce, pesto-drip,...... for those whose keyboard/desk is also their 'feeding station' could be huge. Seriously though - its clever and great to have some put their talent to something with humour instead of lets have the 57th variant of some js thingy. Well done.


Or another goddamn python datetime library


Yes! This would go wonderfully with LobLib, the Lobster library:

https://github.com/bae43/LobLib/


I normally wouldn't pair seafood with coffee, though.

However, a tea stain or wine stain package would indeed pair nicely with LobLib, now wouldn't it?


Love it! Any thoughts on supporting different shades too? I like extra cream.


Lian Tze created this handy variant if you prefer to give a different impression :)

https://www.overleaf.com/latex/examples/latex-wine-stains/cd...


I feel like this package is more appropriate for the humanities than sciences. (I kid!)


They should make the scale of ``lightness'' by creams too; i.e. 1 cream, 2 cream, etc. and by roast as well. At least...that's what I would do if I knew enough LaTeX to write a package.


I'm wondering if these stains would still be realistic if made on a full black background, and an evenly colored background.


How common a use case is that, though? I suppose it's possible someone might print a paper for distribution on paper of some shade other than white, but I've never seen it done.


Slow load but eventually got through.

Here's a direct link to the PDF with examples of the coffee stains provided:

http://hanno-rein.de/downloads/coffee.pdf



Can we have wine stains and blood stains too?


(2009)


Really? Why?

EDIT: I see this terse comment (or rather, a comment with the form of) often, but since the articles are often javascript framework related, I just assume it's because that world keeps shifting. But, here, not so much. I don't see any guidance in the guidelines to denote the age of "timeless" publications. On the other hand, I saw somebody demand/request/suggest "(1912)" after an Edgar Rice Burroughs work some time back. I guess I'm missing something?


It's common to provide the year of publication if it isn't current as some additional metadata. Edit to add: This short, terse comment serves two purposes: lets mods or the submitter know that they may want to update the title, and lets viewers know the year in the meantime. While it's not in the guidelines, it's customary. Currently six of the submissions on the front page feature years. The "[pdf]" and "[video]" "tags" are similar.


Gotcha. That makes sense. Thank you.

> It's common to provide the year of publication if it isn't current as some additional metadata.

I haven't seen that elsewhere (on other sites), except in posting bibliographies, and where it is common, I've seen title-author-publication-date as a bundle. It's not in the HN Guidelines; has this become a community thing? Is it considered (by the community members) helpful?


has this become a community thing? Is it considered (by the community members) helpful?

Yes to both. If it weren't, I doubt members would continue the practice. I find it useful to know that it may be something I've come across in the past. It also puts it in some context.


Have you considered this may be a form of cargo-cult?

This editorializing of the titles is similar to communities that have higher-ups that are equal-but-more-so (mods) that suddenly because of their status decide that their view must be the right one.

Mods adding the (publication year) is exactly like saying "hey, I've kind of read this, it's kind of old and you may (not) be interested in it because it's <years old>. I have no idea of the real value of the text but the year may give you a clue". Please don't.

I don't have any issues with original submitter doing this.


I personally find it useful for the reasons I gave above. Given that other members of the community sometimes complain that submissions aren't new or current, I think it's likely useful to them as well. Submission titles are also changed for other reasons in keeping with the guidelines, such as please use the original title, unless it is misleading or link bait.

https://news.ycombinator.com/newsguidelines.html

Similarly, the "[pdf]" and "[video]" tags are included in the guidelines:

If you submit a link to a video or pdf, please warn us by appending [video] or [pdf] to the title.

It sounds like you don't find it useful, which is fair enough. Do you find it harmful or misleading in some way? If so, how so?

Edit to add: Parent updated with additional rationale:

Mods adding the (publication year) is exactly like saying "hey, I've kind of read this, it's kind of old and you may (not) be interested in it because it's <years old>. I have no idea of the real value of the text but the year may give you a clue".

I think you're reading too much into it. For me, it just puts the post into chronological context. It doesn't give me the impression that it's any less worthwhile to read. In some cases, the year provides additional interest, particularly if it's older.


I understand complaints for "not new or current" if the reader is expecting hacker _news_. But unless the expectations changed too much in last years this site is not necessarily about _news_ as it is about _information_.

If the link is worth it then the year is irrelevant. If it's not worth it, the year won't save it. Thus, adding the year just satisfies our appetite for novelty.


>Thus, adding the year just satisfies our appetite for novelty.

The reverse for me, a link that's still interesting despite the intervening years is one I'm more likely to click.


It often helps to answer "I've seen something with this title before. Is this the thing I know already, or something new?" without clicking the link.


I personallu consider a yearstamp to be a decent disambiguation between "this is current news" v. "this is old news that's still interesting".


Thanks! Updated.


A must have.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: